A methodology for creating semantic digital twin models supported by knowledge graphs

Digital Twin (DT) is a core concept in the digitalization process in the Industry 4.0 and upcoming innovations such as Industry 5.0 and Metaverse. This digitalization can result in a huge amount of data that has not been collected before and now can help to understand and optimize processes and products. However, an important challenge is to really understand what the data can mean and then find out new information and knowledge that can be used for further optimizations. Therefore, it is important to have the virtual representations that the DT proposes as close as possible to the real-world application in a way that it can be understood in the same way by different people and systems. In this context, semantics annotations can be used to organize data by inserting knowledge into the model and providing a common understanding of a specific concept. This paper proposes a methodology for creating semantic digital twin models using the well-known tool called Node-RED. A sequence of steps is described that covers data acquisition, semantics annotations to this incoming data, assets modeling with their relationships which generates a graph of nodes and connections that is saved into a graph database. To be able to realize such a model, 7 so-called nodes have been developed for Node-RED. Finally, a use case has been implemented to demonstrate the application of the proposed methodology.